Peter Mouginis-Mark, Scott Rowland, Harold Garbeil, and Mary MacKay, Hawaii Institute of Geophysics and Planetology, University of Hawaii, Honolulu, HI 96822 USA

1. SUMMARY OF OBJECTIVES

Throughout our investigation, our primary objective has been to learn more about active volcanoes via the analysis of their topography, surface change, and the distribution of flows and cinder cones. To do this, we have used TOPSAR, multi-parameter SIR-C/X-SAR data, and the interferometric analysis of the repeat-orbit data obtained at the end of the SRL-2 mission.

Our main study areas were the Western Galapagos Is. (which formed our volcanology Super Site), and the Big Island of Hawaii. In addition, we have used Piton de la Fournaise on Reunion Island as our third site. We were also alert to the "targets of opportunity" that became available as additional volcanoes were imaged by SIR-C/X-SAR. In this regard, we have given most attention to the analysis of data for Mt. Pinatubo and Taal (both in the Philippines), the Virunga Chain (Zaire), and Erta Ale (Ethiopia).

2. PERSPECTIVES FROM OUR USE OF THE SIR-C/X-SAR DATA

Our investigations of volcanoes with the SIR-C/X-SAR radar has provided us with numerous insights into how best to configure a radar system for future volcano studies. These "insights" can be summarized as follows:

a) There is considerable value in using radar to map volcanoes that are frequently cloud covered (no big surprise). Our studies of volcanoes in the Virunga Chain (Zaire) and the Philippines indicate to us that two-wavelength, two-polarization data are best for mapping due to the trade off between swath width (narrower for quad-pol) and diversity of returns. Figure 1 shows this potential for the infrequently studied Kanlaon volcano, on Negros Island, Philippines.

We can further refine our specification of radar parameters to an incidence angle of 35 - 50° for mapping all types of volcanic features, unless low-relief (< ~15 meter high) structural features such as scarps and flow edges need to be mapped. We also found that a 20 MHz pulse bandwidth was much better at showing subtle differences in lava flow morphology due to the better spatial resolution (Figure 2).

b) For our study, by far the most useful and exciting aspect of the two Shuttle missions was the last four days of the SRL-2 flight, when an exact-repeat orbit was flown to facilitate interferometry experiments. Detecting surface change on Kilauea (Figure 3) and Kliuchevskoi volcanoes provided a concrete example of the value of one day repeat coverage. Very stimulating ideas were also generated for the comparison of repeat-pass data obtained by comparing SRL-1 and SRL-2 observations for Kilauea and Mt. Pinatubo also demonstrate the value of long-term radar interferometry studies, confirming earlier ideas from ERS-1 and ERS-2 data.

Our radar interferometry experiments have indicated that incidence angles in the 40 - 55° range are best, due to the comparative ease of unwrapping the phase data on steep volcanoes. We have worked almost entirely with the L-band interferometry data, since again these data ease the phase unwrapping process. Where baselines are short, we have successfully used the C-band data, but we have been unable to get useable results from X-SAR observations (Figure 4).

c) Prior to SRL-1, we were very uncertain which radar modes would be best for our project. We also continued this lack of knowledge into the SRL-2 mission, due to the short time interval between the flights. As a result, we have ended up with dozens of great images of volcanoes, but nothing like a coherent data base that might be used for inter-mission comparisons or SRL-1/SRL-2 interferometry studies. Only two data takes over volcanoes (over Kilauea and Pinatubo) were obtained with the same modes and incidence angles - all other volcanoes were imaged in slightly different ways, so that we are unable to perform studies for detecting surface change during the intervening months via interferometry. This discovery post-SRL-2 is particularly sad when we consider what we may have seen at the Galapagos Islands in terms of slow deformation on the dormant volcanoes; we collected many data sets for the Galapagos Islands, but in our planning we opted for multi-incidence angle wall-to-wall coverage instead of exploring the interferometry options.

d) Accessing data sets has consistently improved with time. While it was never very slow just after SRL-1, in 1997 we have found the Eros Data Center to be particularly speedy in processing requests. JPL did an outstanding job in 1995 and 1996. We also received excellent support from our X-SAR colleagues throughout this period. Thanks!

3. ACCOMPLISHMENTS

We can summarize the wide range of accomplishments from our project as follows:

1) We have published 16 papers on our investigations of volcanoes in Hawaii, Galapagos, and Zaire using SIR-C/X-SAR and TOPSAR data. In addition, we have given 27 different oral presentations in the U.S., Japan, Australia, and Europe on our SIR-C investigations. Listings of these papers and oral presentations are provided at the end of this report.

2) Field work has been an important part of our validation of the SIR-C/X-SAR data sets. We conducted a 3-week long field expedition to the Western Galapagos Islands in 1989 in preparation for SIR-C. This work was mainly focused on Fernandina volcano, although we also looked at Wolf, Darwin and Alcedo volcanoes. Our Hawaii field work has been an on-going effort, and we have probably made over 15 field trips to the volcano to support our radar studies and other remote sensing investigations. At the start of this investigation, we focused on the interpretation of multi-parameter radar data, but over the last 3 years our attention shifted to the analysis of digital topographic data collected from TOPSAR and SIR-C interferometric data. One of our students, Ms. Selima Siddiqui, went to Taal volcano (Philippines) as part of the 1995 Decades Volcano Workshop. Her Masters thesis is devoted to the analysis of Taal using SIR-C and SPOT data.

3) A highlight in terms of new data sets that we collected was the excellent TOPSAR data set for the Western Galapagos Islands (Fernandina and Isabela Islands). There are six active volcanoes here, but until now their has been very poor knowledge of the topography of these islands. TOPSAR data have enabled us to perform quantitative analyses of the slopes of the volcanoes, the thicknesses of lava flows, and the structure of the volcanoes. We have published numerous papers that use these TOPSAR data (see our bibliography at the end of this report).

4) Always an important issue is the training of the next generation of scientists to use orbital radar data. As part of our SIR-C/X-SAR project, we have graduated 2 Ph.D. and 1 Masters students, with a 2nd Masters degree due for completion this summer.

5) Following extensive help from Howard Zebker (Stanford University) and Paul Rosen (JPL), the University of Hawaii has been able to develop its own ability to process interferometry data. While we are still far from being at the "cutting edge" of development of radar interferometry data processing, our focus has been on the production of specific data products that can aid our volcanological investigations. For instance, we have generated new digital elevation models (DEMs) for 6 volcanoes (Kilauea, Mauna Kea, the Eastern Virunga Chain, Taal, Pinatubo, and Piton de la Fournaise).

6) As part of the Science Team effort, we provided crew training in Hawaii and at JPL/JSC for the SRL-1 mission.

7) Thanks to Jeff Plaut and Tom Farr (both at JPL), we were able to maximize the number of International Decade Volcanoes that were imaged during the two SIR-C/X-SAR flights. 11 out of 13 volcanoes were imaged at least once each.

8) Because of our on-going studies of SIR-C/X-SAR and TOPSAR data, we have been able to provide substantive input into the planning for the next generation of NASA orbital radars. We assisted in both the ECHO proposal and LightSAR planning.

4. ON-GOING & FUTURE STUDIES

It is clear to us that we will be studying aspects of the SIR-C/X-SAR data set for years to come. While we will continue to use the Hawaii data set for numerous education projects (see "Other Uses" below), there are some very interesting volcanology projects that we intend to work on until our funds run out. For instance, we will conduct an analysis of slopes of Piton de la Fournaise as comparison to Galapagos volcanoes. The central cone of Piton de la Fournaise has very steep slopes that we can measure using our SIR-C derived DEM. We want to investigate whether the presence of arcuate fissures and dikes in the walls of the caldera have been responsible for the steeper slopes, which appears to be the case from our analysis of the TOPSAR data for the Galapagos volcanoes.

With the assistance of Dr. Steve Self (University of Hawaii) we plan to investigate the distribution and growth of lahar deposits at Mt. Pinatubo. These are highly dangerous, since the remobilized ash from the June 1991 eruption continues to erode and bury local villages. In addition to data from the two SIR-C/X-SAR flights, we also have several SPOT images taken at different times that will enable us to track the history of these lahar deposits, as well as better understand how the radar data can give us information on the stability of the flanks of the volcano. This work is likely to continue into the coming years since new TOPSAR data were recently (November 1996) collected over Pinatubo as part of the PacRim deployment. We also have on order a large number (10 - 16) ERS-1 and ERS-2 images of Mt. Pinatubo, so that we will study the time variability of the lahar deposits over the time period February 1992 to at least May 1997. In addition, we hope to work with volcanologists from the Philippines in the analysis of the SIR-C data for Kanlaon (Figure 1) as part of the PacRim initiative.

Our current Masters student, Selima Siddiqui, has also shown that a considerable amount of information about the structure and morphology of Taal volcano can be obtained from the SIR-C data. Not only do the multi-parameter SAR images show variations in the degree of erosion of the ignimbrite sheets, but our in-house generated DEM allows us to better study the details of the topography of the volcano. Taal was another PacRim target in 1996, so that we expect to work on this volcano in the coming months.

Structural mapping of Erta Ale volcano, Ethiopia , also appears to be a worthwhile activity using the SIR-C data. One of the key issues here that we will address is the identification of scarps associated with recent faulting of the main volcanic edifice. In SPOT and Large Format Camera images of Erta Ale, it is very hard to see which faults have significant vertical off-sets. These show up clearly in the SIR-C images. Mapping the volcano using both the SPOT and SIR-C data appears to offer exciting new insights into the extensional tectonics associated with this African Rift Zone volcano. An example of the opportunities of using SIR-C data in this area is shown in Figure 5, where L-band measurements are showing the outline of lava flows buried beneath a veneer of sand.

As part of our effort to learn more about processing interferometric SIR-C data, we have searched for surface changes on the flanks of Kliuchevskoi volcano, Kamchatka. This volcano exploded about 5 hours into the SRL-2 mission, and so by the time that the repeat-orbit component of the mission was underway, the eruption had stopped. However, we believe that several of the lahars remained mobile for several additional days, and so may have caused decorrelation of the radar phase data. We plan to conduct a rigorous search of all permutations of the orbit pairs (there were four repeat-orbits over this site where SIR-C data were obtained) as a way of constraining possible lahar movement.

5. OUR OTHER USES OF SIR-C/X-SAR DATA

It is also worth noting that we have used numerous SIR-C images in a variety of educational projects that expose the community to the missions' data, even if the end result is not a research paper. For example, many volcanoes imaged form basis of time series data base for Earth Observing System (EOS) IDS investigations. This will be a multi-year study of volcanoes around the world, using the EOS satellites that start to fly in 1998. Because we were able to collect SIR-C/X-SAR data for numerous volcanoes, including many of the Decade Volcanoes, we have already started the timeline against which we can search for geomorphic change on these volcanoes. This is particularly important on volcanoes such as Mt. Pinatubo, where very active mudflows ("lahars") continue to change the landscape on an annual basis following the 1991 eruptions.

One of the other Geology Super Sites was Death Valley. While we have no research interest here, we have used these data as the central part of our undergraduate remote sensing field trip to California. By showing the students the different types of radar data for a geologic landscape that is not volcanic (all that we have in Hawaii), we are better able to teach them the uses of remote sensing data. About 30 undergraduate students have worked with these SIR-C data in the field.

The Hawaii Space Grant Consortium runs a Web site called "Virtually Hawaii". This site receives ~2.7M hits per month from over 65,000 different users. On numerous pages, we have placed SIR-C data of the Big Island, Maui and Honolulu as a way of showing people what Hawaii looks like at different wavelengths. We feel that this is a great way to help promote the general public's interest in radar data.

Analysis of our Kilauea volcano SIR-C site enabled us to collect a large variety of radar and other remote sensing data sets of the area. The set of data includes AIRSAR, TOPSAR, SPOT, Landsat, TIMS and air photography. This is a great teaching resource, that we have used not only for our own classes, but also at the national level. As part of NASA's Solar System Exploration Division's Planetary Geology and Geophysics Program, we have run three week-long field workshops on the volcano. We have used the SIR-C/X-SAR data as an analog to the Magellan data for Venus, as well as a way to better interpret other volcanoes on Earth. About 40 graduate students, post-docs and principal investigators in the PG&G Program have taken these workshops and used SIR-C data for Kilauea volcano.

6. Bibliography

Campbell, B.C., Zisk, S.H., and Mouginis-Mark, P.J. (1990). Interpreting lava surface textures from quad-pol radar data: An inverse model and comparison with topography measurements. Remote Sensing of Environment, 30, 227 - 237.

Evans, D. L., T. G. Farr, H. A. Zebker, J. J. van Zyl, and P. J. Mouginis-Mark (1992). Radar interferometric studies of the Earth's topography. Eos, 73, p. 553 and 557 - 558.

Francis, P.W., G. Wadge, and P.J. Mouginis-Mark (1996). Satellite monitoring of volcanoes. In: Monitoring and Mitigation of Volcano Hazards, R. Tilling and R. Scarpa (Eds.), Springer-Verlag Press, N.Y., pp. 257 - 298.

MacKay, M. E. and P. J. Mouginis-Mark (1997). The effect of varying acquisition parameters on the interpretation of SIR-C radar data: The Virunga Volcanic Chain. Remote Sensing of Environment, 59, 321 - 336.

Mouginis-Mark, P.J. (1994). Mitigating volcanic hazards through radar interferometry. Geotimes, p. 11 - 13, July 1994.

Mouginis-Mark, P.J. (1995). Preliminary observations of volcanoes with the SIR-C/X-SAR radar. IEEE Trans. Geosci. Rem. Sen., 33, 934 - 941.

Mouginis-Mark, P. J. and Francis, P.W. (1992). Satellite observations of active volcanoes: Prospects for the 1990's. EPISODES, 15, 46 - 55.

Mouginis-Mark, P. J. and H. Garbeil (1993). Digital Topography of Volcanoes from Radar Interferometry: An Example from Mt. Vesuvius, Italy. Bull. Volcanol., 55, 566 - 570.

Mouginis-Mark, P. J., S. K. Rowland and H. Garbeil (1996). Slopes of western Galapagos volcanoes from airborne interferometric radar. Geophys. Res. Lttrs., 23, 3767 - 3770.

Munro, D.C. and Mouginis-Mark, P.J. (1990). Eruptive patterns and structure of Isla Fernandina, Galapagos Islands, from SPOT-1 and Large Format Camera images. Int. J. Rem. Sen., 11, 1501 - 1509.

Munro, D.C. and S.K. Rowland (1996). Caldera morphology in the western Galapagos and implications for volcano eruptive behavior mechanisms of caldera formation. J. Volcanol. Geotherm. Res., 72, 85 - 100.

Munro, D. C., Rowland, S. K., Mouginis-Mark, P. J., Wilson, L., and Oviedo-Perez, V.-H. (1991). An investigation of the distribution of eruptive products on the shield volcanoes of the western Galapagos Islands using remotely sensed data. Submitted to: Eighth Thematic Conference on Geologic Remote Sensing, Denver, CO, April 29 - May 2, 1991, p. 1161 - 1174.

Rowland, S.K. (1996). Slopes, lava flow volumes, and vent distributions on Volcan Fernandina, Galapagos Islands. J. Geophys. Res., 101, 27657 - 27672.

Rowland, S.K. and D.C. Munro (1992). The caldera of Volcan Fernandina: A remote sensing study of its structure and recent activity. Bull. Volcanol. 55, 97 - 109.

Rowland, S.K., D.C. Munro, and V.H. Perez-Oviedo (1994). Volcan Ecuador, Galapagos Islands: Erosion as a possible mechanism for the generation of steep-sided basaltic volcanoes. Bull. Volcanol. 57, 117 - 126.

Zebker, H. A., P. Rosen, S. Hensley, and P. J. Mouginis-Mark (1996). Analysis of active lava flows on Kilauea volcano, Hawaii, using SIR-C radar correlation measurements. Geology, 24, 495 - 498.

7. ORAL PRESENTATIONS ABOUT SIR-C/X-SAR

March 15th 1991: "Remote Sensing of Volcanic Processes and Landforms". Coastal Studies Institute Seminar, Louisiana State University.

March 30th 1991: "Remote Sensing of Volcanoes". Hawaii Volcano Observatory Seminar.

May 4th 1991: "Satellite observations of active volcanoes: Interdisciplinary investigations, EOS and other satellites". IUGS Workshop on Remote Sensing in Global Geoscience Processes. Boulder, Colorado.

August 1st 1991: "Remote Sensing of Active Volcanoes", Geophysical Institute Seminar, University of Alaska.

August 26th 1991: "Remote Sensing of Volcanoes: SIR-C, ERS-1 and EOS", National Air and Space Museum Seminar, Smithsonian Institution, Washington, D.C.

May 27th, 1992: "Radar Studies in Physical Volcanology", IGARSS International Conference, Houston, Texas.

May 27th, 1992: "SIR-C/X-SAR Experiments in Physical Volcanology: Galapagos, Hawaii and Reunion Island", Briefing to Space Shuttle Astronauts, NASA Johnson Space Center, Houston, Texas.

June 1st, 1992: "Current Trends in the Remote Sensing of Volcanoes: The Earth Observing System and Other Missions", Visiting Lecture Series, U.S.G.S. Menlo Park, CA.

May 19th, 1993: "Recent Progress in Monitoring Volcanoes and Volcanic Hazards with Satellite and Airborne Remote Sensing", Geophysical Institute Seminar Univ. Alaska.

May 26th, 1993: "Use of TOPSAR Digital Topographic Data for Volcano Studies: Geometry of Valleys on Mt. Somma, Italy". Spring 1993 AGU Meeting, Baltimore, MD.

September 29, 1993: "Use of Digital Elevation Models for Analysis of Volcanic Hazards and Surface Processes", IAVCEI Congress, Canberra, Australia.

October 28, 1993: "Current and Future Uses of TOPSAR Digital Topographic Data for Volcanological Research", 4th JPL Aircraft Geosciences Workshop, Washington, DC.

November 3, 1993: "SIR-C/X-SAR: Galapagos Islands, Hawaii and Objectives for EOS" SIR-C/X-SAR Team Meeting, Rome, Italy.

November 17, 1993: "Remote Sensing of Volcanic Hazards", Joint U.S./Japan Workshop on the Application of Remote Sensing Technology to Disaster Reduction, Tsukuba, Japan.

February 16th, 1994: "The SIR-C/X-SAR Volcanology Experiments", presentation to Dr. Charles Kennel, NASA MTPE Associate Administrator, Washington, DC.

May 20th, 1994: "Monitoring Volcanoes in Alaska with Satellites", USGS Alaska Volcano Observatory

June 14th, 1994: "Radar Studies of Hawaii", USGS Hawaiian Volcano Observatory.

October 24th, 1994: "Space Shuttle Radar Observations of Volcanoes", Geological Society of America Annual Meeting, Seattle.

December 2nd, 1994: "Analysis of volcanic and earthquake hazards using radar interferometry", 1st Workshop on SAR Interferometry, Tokyo, Japan.

January 9th - 11th, 1995: "Future of U.S. Radars in Geology and Topographic Mapping", NRC Review, Irvine, CA.

July 5th, 1995: "Volcanology Applications of Radar Interferometry", invited talk All-Union Session, IUGG General Assembly, Boulder Colorado.

July 13th, 1995: "NASA Global Change Studies", guest lecture at the Department Earth Sciences, Open University, Milton Keynes, England.

February 14th, 1996: "Recent Results from Volcano Studies using the SIR-C/X-SAR". SIR-C Team Meeting, Santa Barbara, CA.

April 2nd, 1996: "Radar and Topographic Studies", 1st U.S.-Japan Earth Remote Sensing Conference, Kona, Hawaii.

July 8th, 1996: "Perspectives on Earth Remote Sensing in the United States", CSIRO Divn. of Atmospheric Chemistry, Melbourne, Australia.

July 9th, 1996: "Radar and Field Studies of Volcanoes in the Galapagos Islands", Geology Department Seminar, Monash University, Melbourne, Australia.

July 22nd, 1996: "Remote Sensing of Volcanoes", NE Australia Chapter of IEEE, Townsville, Queensland, Australia.

March 12th, 1997: "The Eruptive Styles of Basaltic Shield Volcanoes from Shuttle Imaging Radar-C (SIR-C) and X-SAR Data", SIR-C Team Meeting, Florence, Italy.

March 26th, 1997: "TOPSAR Digital Topography for the Analysis of Pacific Rim Volcanic Hazards", PacRim Team Meeting, Monrovia, CA.

March 27th, 1997: Validation of TOPSAR Data for Kilauea Volcano, Hawaii", PacRim Team Meeting, Monrovia, CA.

Figure Captions:

Figure 1: SRL-2 image of Kanlaon volcano on Negros Island, Philippines. This volcano is under intense study by local volcanologists, since the volcano has been highly active in recent years, and last erupted in 1996. Data Take 110.30. Red (L-HH), green (L-HV), blue (C-HV).

Figure 2: Comparison of SRL-1 data for the north flank on Nyamuragira volcano, Zaire, showing the difference in spatial resolution between 10 MHz (SRL-1 DT 58.61) and 20 MHz (SRL-1 DT 171.1) pulse band widths. The panels at right show the same area as seen by the SPOT satellite (10 m/pixel). See also MacKay and Mouginis-Mark (1997) for details.

Figure 3: Zebker et al. (1996) describe the exciting application of the repeat-orbit SRL-2 data for the analysis of the growth of the lava flow field at Kilauea volcano, Hawaii. The top image shows the location of the study area in relation to the Pu'u O'o and Kupaianaha vents, and the Royal Gardens subdivision (virtually destroyed between 1983 and 1986). The middle panels show field photographs taken on October 9th, 1994 during the SRL-2 mission. Red arrows identify then-active flows which cause complete decorrelation in the three SIR-C correlation maps (lower three panels) on the four days when repeat-orbit data were obtained. SRL-2 data takes 122.4, 138.4, 154.4, and 170.4 were used to investigate these surface changes.

Figure 4: We have experimented a lot with the repeat-orbit coverage of Mauna Kea, Hawaii, from the SRL-2 mission. here we compare data collected on October 8th and 9th, 1994, to show the difference in the ease in interpreting the phase cycles for the volcano. Also shown are an L-band VV backscatter image (top left) and the L-band phase correlation image (to right). SRL-2 data takes 138.4 and 154.4 were used in this analysis.

Figure 5: SIR-C coverage of Erta Ale volcano in NE Africa is tantalizing, but we did not collect repeat-orbit data so that no digital elevation model could be produced. This image shows how a thin veneer of sand has partially buried a lava flow on the southern flanks of the volcano. SRL-2 Data Take 34.52. 52° incidence angle, mode 11. Red (L-HH), green (L-HV), blue (C-HV).